2980 results about "Target surface" patented technology

The present application relates to a weld seam surface defect detection method and system. The weld seam surface defect detection method comprises: performing multi-angle and multi-surface acquisition of target surface images; then, reconstructing normal maps of the target surface images by means of a high-performance algorithm, and clarifying normal features of defects; and finally, on the basis of the surface normal features, aiming at transverse texture defects and height mutation defects, respectively designing a transverse texture defect detection algorithm based on two stages and rotated rectangle extraction and a height mutation defect detection algorithm based on adaptive dual-threshold segmentation. The present application overcomes texture noise and pseudo-defect interference of ground surfaces, effectively solves the problems of difficulty in feature extraction and positioning of ground surfaces of steel rail weld seams, and has the system operating efficiency superior to that of existing detection methods, thus satisfying requirements of actual production.

An illustrative system may include a stereoscopic imaging device comprising: a first image sensor having a first pixel array and configured to provide a first windowed output comprising first image information associated with pixels included within a first window of the first pixel array, the first window having a first position; and a second image sensor having a second pixel array and configured to provide a second windowed output comprising second image information associated with pixels included within a second window of the second pixel array, the second window having a second position; and an imaging control system that may be configured to: determine a working distance between a target surface within a scene and the stereoscopic imaging device; and adjust one or more of the first position of the first window within the first pixel array or the second position of the second window within the second pixel array.

The invention relates to the technical field of surface detection, and discloses a surface detection method and system for a precision fastener, and the method comprises the steps: carrying out the multi-angle image collection and image preprocessing of the precision fastener, and obtaining an initial feature image set; three-domain feature extraction is carried out, and surface feature fingerprint vectors are obtained through calculation of a texture feature domain, a shape feature domain and a material feature domain; constructing a depth separable convolution structure and a feature pyramid structure to obtain a depth surface detection network; taking the pre-trained target detection model as a teacher network, taking the deep surface detection network as a student network, and obtaining a target surface detection model through calculation of a characteristic distillation loss function and a response distillation loss function; and inputting a to-be-detected precise fastener image into the target surface detection model, and obtaining category information and position information of surface defects through two-stage measurement recognition calculation. According to the method, the calculation efficiency is improved while the detection precision of the precise fastener is ensured.

The invention relates to a remote sensing image ground feature recognition method based on deep learning, and the method comprises the steps: carrying out the data collection and preprocessing of a ground surface target region, eliminating the position deviation through geometric correction, processing the illumination difference through combination with radiation equalization, and generating a ground feature registration image; performing multi-dimensional feature fusion processing on the image, performing tensor fusion on vegetation spectral features, earth surface texture features, point cloud data features and linear ground feature features, and constructing a ground feature fusion matrix; a bilateral convolutional neural network is adopted to extract spectral response characteristics and spatial correlation characteristics, and characteristic interaction is realized through an attention mechanism to generate a ground feature probability distribution diagram; and finally, carrying out noise filtering, boundary refining and vectorization conversion processing on a classification result, and outputting a ground feature classification vector diagram. According to the method, three technical bottlenecks of insufficient cooperative utilization of multi-source heterogeneous data, insufficient spectrum-space feature fusion and poor GIS compatibility are solved, and the operation efficiency of territorial investigation, disaster monitoring and other scenes can be remarkably improved.

The invention discloses an unmanned aerial vehicle-mechanical arm system cooperative control method for precise spraying. According to the method, through a depth camera and a point cloud reconstruction algorithm, three-dimensional modeling of a target surface and semantic recognition of a spraying area are achieved, and a spraying path and spraying posture data are generated. Based on spraying task requirements, a spraying film thickness experience estimation model is established, and path point-level spraying parameters are obtained. Load mass change, mechanical arm mass center change and spraying reaction force of each path point are calculated, and a coupling dynamic model is constructed in combination with the system state. And dynamic compensation control input is generated by using an adaptive sliding mode control algorithm. After spraying is completed, the actual spraying effect is collected through image and point cloud detection, color, film thickness and texture features are extracted and compared with task requirements, error feedback is generated, and control parameters are adjusted online. According to the method, the problems of dynamic change and error compensation in the spraying process are effectively solved, and the method is suitable for a high-precision spraying task in a complex environment.

The invention discloses an error compensation inductance sensor calibration method and system, and the method comprises the steps: controlling a macro-micro composite driving platform to drive a moving part provided with a to-be-calibrated inductance sensor, and enabling the moving part to carry out the continuous displacement movement relative to a fixed standard target surface; measuring the actual displacement of the moving part to obtain a standard displacement sequence; the actual output voltage of the inductive sensor to be calibrated in the movement process is synchronously collected, and an actual measurement value sequence is formed; solving measurement deviation, generating a measurement deviation sequence, and dividing the sequence into not less than three compensation intervals according to a displacement range; a compensation value lookup table is established in each compensation interval, and calibration parameters used for correcting zero offset, gain drift and nonlinear characteristics are generated. Displacement measurement and sensor output synchronous acquisition are combined, zero offset, gain drift and non-linear errors can be corrected in a targeted mode, and compensation fineness and adaptability are improved.

The invention relates to the technical field of intelligent target scoring, in particular to an intelligent individual soldier target scoring system and method, and the system comprises a target surface calibration module, a hit recognition module, an offset mapping module, a mode switching module and a terminal output module. According to the method, the thermal expansion trend is analyzed by collecting the temperature data of the target surface sensor, the target surface can be calibrated more accurately, impact point positioning can be corrected, shooting target positioning can be optimized, the accuracy of the shooting result can be enhanced, hit points which are not consistent with the preset angle can be screened out by recording the stress vibration and ballistic characteristics of the target surface during shooting, and the accuracy of the shooting result can be improved. The authenticity and reliability of shooting data are effectively improved, the gun holding posture change of a shooter in continuous shooting is monitored, the posture change is associated with hit offset, the shooting posture is helped to be adjusted, the overall shooting precision is improved, the shooting mode is dynamically adjusted, the shooter can better adapt to different shooting environments, and the shooting efficiency is improved. And the pertinence and the effect of training are obviously improved.

The invention discloses an automobile part surface defect detection method based on a large model. The method comprises the following steps: acquiring feature data of a target surface by using a multi-dimensional imaging device; feature extraction, clustering analysis and abnormal mode recognition are carried out through the large model; extracting an abnormal region and generating a statistical model; comparing the statistical model with standard sample data to determine the position of an abnormal region; defect types are classified and identified, and resource allocation is dynamically optimized and calculated to adapt to detection complexity; and finally outputting defect types and positions. According to the method, the spectral imaging technology and the deep learning algorithm are combined, high-precision detection, positioning and classification of defects such as cracks, scratches and bubbles are achieved, a visual detection report is generated, the method is suitable for real-time quality control in large-scale production, the detection efficiency and accuracy are improved, and manual errors and cost are reduced.

The invention relates to the technical field of dispensing positioning, in particular to a trinocular vision-based five-axis dispensing machine and a precision calibration method thereof.The dispensing machine comprises an X-axis moving mechanism, a Y-axis moving mechanism, a Z-axis moving mechanism, an AC rotary workbench and a dispensing device, the Z-axis moving mechanism is provided with a first shooting assembly and a second shooting assembly, and the Z-axis moving mechanism is provided with a third shooting assembly and a fourth shooting assembly; a first calibration plate, a third shooting assembly and a second calibration plate are arranged on the AC rotary workbench; the precision calibration method comprises the following steps: S1, installing a camera; s2, camera self-calibration; s3, performing visual measurement on an X axis, a Y axis, a Z axis, an A axis and a C axis; s4, performing visual calibration; s5, comprehensive compensation is carried out, and the correction compensation effect is verified through real-time error correction and track retest; and S6, carrying out global iteration optimization strategy on the corrected and compensated actual pose model based on an LM algorithm. According to the invention, the stability of calibration precision can be ensured, the matching degree of a dispensing path and a target surface is improved, and high-precision dispensing is realized.

Implementations disclosed describe an inspection device capable of being transferred by a robot blade into a processing chamber of a manufacturing machine, the inspection device comprising an optical sensor to detect light reflected from a target located within the processing chamber, wherein the optical sensor is to output, to a processing device, a signal representative of a state of a region of a surface of the target.

The invention discloses a projectile drop point detection method, system and equipment based on dual-light end-to-end fusion, and a medium. The method comprises the following steps: collecting visible light imaging data and thermal infrared imaging data of all target regions; preprocessing the data to obtain visible light and thermal infrared images of corresponding target region time-space registration; projecting a ground target area on the obtained image after space-time registration to a corresponding reference projection target surface; inputting the registered image into a multi-modal fusion detection model, and judging whether an explosion phenomenon is detected in the image frame by frame to obtain a bounding box corresponding to an explosion target in the image; according to the explosion bounding box, resolving and correcting the position of the drop point by adopting a drop point resolving algorithm, and determining the image coordinate of the drop point in the explosion area; and mapping the drop point to a reference projection target surface, obtaining position coordinate information of the drop point in the reference projection target surface, and recording and outputting relative position information of the drop point on the reference projection target surface. The detection precision is improved, and the target scoring task is completed.

The invention provides a passive radar seeker anti-interference identification method based on a dual-polarization array, and the method comprises the steps: obtaining a first signal and a second signal, and carrying out the time-space alignment calibration of the first signal and the second signal, and obtaining a first polarization signal and a second polarization signal; performing polarization-Doppler-frequency band three-dimensional tensor construction on the first polarization signal and the second polarization signal; performing dynamic core tensor decomposition processing on the polarization-Doppler-frequency band three-dimensional tensor to obtain a target polarization characteristic parameter and an interference polarization characteristic parameter; performing interference suppression processing on the first polarization signal and the second polarization signal through polarization matched filtering to generate an anti-interference enhanced signal; performing polarization scattering mechanism decomposition processing on the target polarization characteristic parameter to generate a target surface scattering component and a dihedral angle scattering component; and outputting an anti-interference identification result based on the target surface scattering component, the dihedral angle scattering component and the polarization phase consistency parameter of the anti-interference enhancement signal. The method can improve the anti-interference recognition precision of the radar.

The invention relates to a 3D size measurement method suitable for a curved surface structure. Capturing a target curved surface by using a camera and a laser scanner; using structured light to generate point cloud data; performing point cloud registration among the plurality of views by means of a SIFT feature matching algorithm to generate a unified point cloud model; using a deep learning algorithm to train the feature recognition model, and extracting key geometric features of the curved surface; identifying curved surface features with different edges, angles and curvatures by marking and classifying, and converting the features into mathematical representation by using a Bezier curve parameterized modeling method; establishing a comprehensive error model, evaluating a plurality of error sources, and fitting an error propagation function through experimental data; adjusting dimension measurement by using a least square method; a real-time data acquisition and feedback system is introduced, environment change is monitored by using a temperature and humidity sensor, and scanning parameters including scanning speed and light source brightness are adjusted through a PID controller.

The invention discloses a surface runoff cause identification method and device based on deep learning and interpretability analysis, and is used for solving the technical problem of low identification precision caused by the existing surface runoff cause identification method. The method comprises the following steps: acquiring meteorological and hydrological factor time series data, and inputting the meteorological and hydrological factor time series data into a surface runoff cause identification model based on deep learning; the surface runoff cause identification model based on deep learning comprises a one-dimensional convolution module, a bidirectional long short-term memory network and an identification module based on a multi-head attention mechanism. A one-dimensional convolution module is adopted to carry out convolution operation on the meteorological and hydrological factor time sequence data, and a local response feature sequence is output; performing bidirectional processing operation on the local response feature sequence through a bidirectional long-short-term memory network to generate a target comprehensive feature sequence; and inputting the target comprehensive feature sequence into an identification module based on a multi-head attention mechanism for identification, and outputting a target surface runoff cause identification result.

The invention relates to the technical field of robot control, and discloses a robot scanning path adjusting method, device, equipment and medium, and the method comprises the steps: controlling a robot equipped with a sensor to move along a preliminary path, collecting a preliminary two-dimensional image of a target object, processing the image, extracting the feature points of the target object, and combining the spatial information of the sensor when the image is collected; the method comprises the steps of generating a primary three-dimensional point cloud, determining a plurality of ideal sensor postures based on the primary three-dimensional point cloud, determining a new sensor position of each posture on a contactable surface to generate a re-planned scanning path, controlling a robot to move along the re-planned scanning path, adjusting the sensor postures, and collecting a final two-dimensional image of a target object. According to the method, the scanning path is optimized in combination with the initial three-dimensional point cloud local surface normal direction, so that the attitude of the sensor is matched with the geometrical characteristics of the target surface, the problem of three-dimensional reconstruction surface breakage caused by inter-frame pose mismatching under free breathing is solved, and the accuracy of autonomous ultrasonic scanning of the robot is improved.

The invention discloses a solar cell and a manufacturing method thereof, relates to the technical field of photovoltaic, and aims to increase the conductivity and effective doping concentration of a first part, covering at least part of the top of a pyramid-like pyramid, of a first semiconductor layer and facilitate reduction of transmission loss of carriers collected in the first semiconductor layer transmitted to a conductive material. The solar cell includes a semiconductor substrate and a first semiconductor layer. The semiconductor substrate comprises a first surface and a second surface which are opposite. At least one of the first surface and the second surface is a target surface. The surface of at least partial area of the target surface is a suede, and a plurality of similar pyramids are formed on the suede. The plurality of pyramids comprise a plurality of first pyramids, the first semiconductor layer is arranged on the textured surface, and the material of the first semiconductor layer comprises amorphous silicon and / or nano-crystalline silicon. The first semiconductor layer comprises a first part covering the top of the first-type pyramid and a second part covering the base of the first-type pyramid, and the crystallization degree of the first part is larger than that of the second part.

The invention relates to an optical fiber coupling dynamic interferometer based on arrayed waveguide grating light splitting. A luminous source and an arrayed waveguide grating of the optical fiber coupling dynamic interferometer are arranged on a motion module. The light emitting source is connected to the input port of the array waveguide grating through an optical fiber, and through the movement module, any one of the multiple output ports of the array waveguide grating is in butt joint with the optical fiber connected to the incident end of the collimation structure. A half-wave plate and a beam splitter prism of the imaging element group are sequentially arranged on an emergent light path of the emergent end of the collimation structure, and the beam splitter prism is used for splitting a light beam passing through the half-wave plate into a reference light beam transmitted along a first light path and a measurement light beam transmitted along a second light path. And the reflecting mirror of the imaging element group reflects the reference light beam back to the beam splitting prism. And the to-be-measured surface reflects the measurement light beam back to the beam splitter prism. The reference beam and the measurement beam reflected back to the beam splitter prism interfere with each other and are imaged on the target surface of the camera. Accurate measurement of the wavefront image of the surface to be measured is realized.

The invention discloses a crack image processing method based on a neural network and plane geometry image transformation, and particularly relates to the field of general image data processing driven by plane geometry image transformation and neural network identification. Comprising the steps that an image collection platform projects a structured image coding pattern to the surface of a to-be-measured target through a preset image coding pattern template, and synchronously collects an image frame containing an image mark so as to construct a space anchor point used for image geometric reference in an image plane; view angle parameter change information in an image acquisition process is acquired through an image acquisition platform, and projection configuration parameters of an image coding pattern are combined. By constructing an image offset modeling mechanism jointly driven by an image coding pattern and a space anchor point, image in-plane correction of geometric distortion features in an image frame is realized, and structural consistency and texture continuity before subsequent image analysis are improved.

The invention discloses a four-channel fluorescence detection light path module of a super-resolution biological microscope, which divides a fluorescence signal of an intermediate image plane (5) into four conjugate channels (7a-7d) through multi-stage dichroscope light splitting, reflector guiding and lens focal plane adjustment on the premise of not introducing additional aberration, and realizes real-time synchronous detection by using a single large-target-surface camera (8). By adjusting the distance between the reflectors, the inclination angle of the parallel plate glass or the position of the lens, the device can adapt to different camera target surface sizes and compensate chromatic aberration, and the problems of motor switching delay and dual-channel aberration in the prior art are solved.

The invention relates to a method for rapidly calculating the RCS of a coated unmanned aerial vehicle under a complex background, and belongs to the field of stealth target electromagnetic scattering calculation. Comprising the following steps: determining reflection coefficients of a single-layer wave-absorbing material with a PEC substrate at different incident angles according to an equivalent transmission line theory; calculating induced current generated by incident electromagnetic waves on the surface of the target coated with the single-layer wave-absorbing material according to the reflection coefficient; based on the induced current, utilizing physical optics and physical diffraction theories to calculate scattered fields of all the illuminated surface elements, and superposing all the calculated scattered fields; the scattering field of each surface element on the sea surface model at a fixed angle is obtained through vector superposition of the scattering fields obtained through an EPO'field and a TSM; and finally, superposing all the scattered fields, and calculating the total field RCS of the complex sea background and unmanned aerial vehicle composite model based on the superposed total scattered field. According to the method, the RCS of the coating unmanned aerial vehicle under the complex background can be quickly calculated.

The invention discloses a polarization spectrum modulation type multispectral stereo imaging method which solves the problem that in the prior art, an existing polarization spectrum imaging scheme does not achieve three-dimensional reconstruction of a target and achieves the purpose that the three-dimensional reconstruction of the target is not achieved under the condition that active illumination or special marks are not needed. Synchronous high-precision reconstruction of physical attributes and three-dimensional geometry of the surface of the target is realized; the method comprises the following steps: uniformly modulating polarization information and spectral information of incident light of a target through a three-way optical filter and a polaroid to obtain modulated incident light information; performing aliasing acquisition of a plurality of polarization spectrum components on the modulated incident light information through a multi-color detector combination to obtain aliasing light intensity; performing linear solution on the aliasing light intensity to obtain each polarization spectrum component vector, calculating the polarization degree and the polarization angle of each channel, and further deriving a surface zenith angle and an azimuth angle; reconstructing the three-dimensional morphology of each spectrum channel through normal gradient field frequency domain integration; and finally, fusing the multi-spectral three-dimensional data to obtain a multi-spectral three-dimensional imaging result.

The invention provides a large-depth underwater target salvaging device and salvaging method, the salvaging device comprises a carrier, a holding claw device, an adsorption mechanism, a jet device and a jet pump; the holding claw device is of an under-actuated structure and is used for clamping a target and providing supporting force from the lower portion and / or the lateral lower portion of the target. The adsorption mechanism adopts an under-actuated structure and is used for adsorbing a target and providing adsorption force from the upper part and / or the lateral upper part of the target; high-pressure water flow is generated at the output end of the jet device through the jetting pump, so that negative pressure is generated in the adsorption mechanism to achieve the adsorption function. Wherein in the process that the holding claw device clamps the target, whether clamping is completed or not is judged according to whether the pressure between the holding claw device and the target reaches a preset value or not; and the vacuum adsorption mechanism performs vacuum adsorption on the target surface after the holding claw device completes clamping. The carrier, the holding claw device and the adsorption mechanism are combined, large-depth underwater targets can be effectively and rapidly fished through driving of the carrier, and the weight capable of being fished is greatly increased.

The invention relates to the technical field of robots and intelligent control, and discloses a robot trajectory control method, device and equipment based on multiple modalities and a medium, and the method comprises the steps: obtaining a natural language instruction, environment visual information and current state information of a robot, carrying out the fusion of visual and language features, and generating a fusion feature, generating a track sequence through a track generation model according to the fusion features and the robot state information, collecting a contact force signal of the tail end of the robot in contact with the target surface, adaptively adjusting impedance parameters by using the contact force signal, and generating a control instruction; and based on the contact force signal, feedback correction is carried out on the track generation model to drive the tail end of the robot to move, and multi-mode sensing and force control feedback cooperative control is achieved. According to the invention, track and force control cooperation is realized through multi-modal fusion and contact force feedback, so that the robot keeps stable fitting and adaptive motion in a complex contact scene, and the track generation precision and the imaging quality are improved.

A portable, modular handheld imaging system is disclosed. The modular system comprises a first housing portion and a second housing portion. The first housing portion includes at least one excitation light source. A first filter is configured to detect and permit passage of selected optical signals responsive to illumination with the excitation light to a first image sensor. A second filter is configured to detect and permit passage of selected optical signals responsive to illumination of the target surface with white light to a second image sensor. The second housing portion is configured to releasably receive the first housing portion. The second housing portion includes a display and a processor configured to receive the detected fluorescent and white light optical signals and to output a representation of the target surface to the display based on the detected optical signals.

The invention discloses a solar cell, a photovoltaic module and a semiconductor substrate, and relates to the technical field of photovoltaics, so that a passivation layer has a good passivation effect on the semiconductor substrate, and the contact performance of a doped semiconductor layer and an electrode and the shaping of the electrode are improved. The solar cell includes a semiconductor substrate and a passivation layer. At least one of the first surface and the second surface of the semiconductor substrate is a target surface. A part of the surface of one side, departing from the semiconductor substrate, of the passivation layer is provided with a plurality of tower-footing-like texture structures which are recessed towards one side close to the semiconductor substrate. The plurality of tower footing-like texture structures comprise a first tower footing-like texture structure and a second tower footing-like texture structure. All corners in the bottom surface of the first-class tower footing texture structure are round-like corners, part of the corners included in the bottom surface of the second-class tower footing texture structure are round-like corners, and the other corners are sharp-like corners. And in the bottom surface of the same second-class tower footing-shaped texture structure, the number of the similar round corners is greater than or equal to the number of the similar sharp corners.

The invention discloses a submarine topography reconstruction system and method based on optical mobile measuring and scanning, and relates to the technical field of ocean exploration, and the system comprises the following units: an image collection unit, which is used for coordinating the emission of 532nm pulse laser and the collection of a gating camera image through a synchronous sequential circuit, a triangulation measurement optical structure with separated emission and reception is adopted, and a submersible can be carried for mobile measurement and scanning; the system calibration unit is used for carrying out joint calibration on camera parameters and a laser plane so as to determine system calibration parameters, so that the spatial position of a laser stripe accurately corresponds to the actual coordinate of the target surface; and the image processing unit is used for preprocessing the image acquired by the system and further extracting the coordinates of the center line of the laser stripe in the preprocessed image. According to the invention, adjacent splicing can be carried out through continuously collected laser stripe images to construct two-dimensional submarine landform information, and three-dimensional reconstruction of submarine topography can be carried out by using generated point cloud data.

The invention relates to the technical field of X-ray tube control, and discloses a control method and device for a controllable flying focus X-ray tube, and the method comprises the steps: building an orthogonal electromagnetic deflection system based on an electron beam flight path of the X-ray tube, and the orthogonal electromagnetic deflection system comprises an X-direction electromagnetic coil and a Y-direction electromagnetic coil; creating a conversion relation for converting the target focus position function into an electromagnetic coil current value; executing nonlinear magnetic field response compensation to obtain high-speed switching control signals of the X-direction electromagnetic coil and the Y-direction electromagnetic coil; and performing focus trajectory prediction based on the high-speed switching control signal to obtain a focus motion trajectory formed by the electron beam on the target material, and performing thermal load distribution optimization on the focus motion trajectory to obtain a focus position control scheme for uniform distribution of the surface temperature of the target material. Accurate positioning control of the electron beam on the target material is achieved, and the X-ray imaging quality is improved.

An apparatus includes a first light source for illuminating a chart including an index surface on which an index is provided, an objective lens for guiding index light emitted from the chart to an optical system, a first element for receiving the index light reflected on an optical surface in the optical system, an interferometer configured to include a second light source and a second element, and acquire a wavefront of measurement light, the measurement light being test light reflected on the target surface, and a reduction unit for reducing an intensity of a signal of light reflected on an optical surface other than the target surface, of the test light reflected on the plurality of optical surfaces and received by the second element.

The invention discloses a fused quartz high-precision surface laser double-sided polishing method, and belongs to the technical field of optical element laser processing. In order to solve the problem of surface shape distortion of the fused quartz element under the existing laser polishing process, the method comprises the following steps: firstly, ultrasonically cleaning the fused quartz element and blow-drying the fused quartz element by using compressed air; the morphology distribution of the fused quartz after laser polishing is simulated through finite element simulation, and the scanning speed is optimized; polishing the target surface according to the process parameters and the processing path after the scanning speed is optimized; the internal temperature distribution of the fused quartz after laser polishing is obtained through finite element simulation, and the polishing power of elements with different thicknesses is optimized; the back face of the target face is polished through the optimized laser power; and after laser polishing, carrying out ultrasonic cleaning and drying again. The method is simple in process and low in cost, the surface shape quality can be remarkably improved, the roughness is reduced, the time cost consumed by a parameter cross experiment is avoided, and optimization of laser polishing parameters is efficiently achieved.